Medicago truncatula

Current Biology Vol 14 No 5 R180

then modeling the interaction of those parts based on their individual properties will fail to capture the larger system dynamics. Over 200 years ago, Immanuel Kant noted this approach would not be suitable for understanding complex systems such as an organism. He said there would never be a “Newton of the blade of grass”: because the blade of grass is more than the sum of its parts, complete knowledge of the parts list will not necessarily be predictive of the behavior of the whole. Fortunately, we don’t have to resort to a vitalistic explanation for the phenomenon of emergence. Systems biology offers both an epistemological and technological solution to this problem. Ideker et al. have outlined the four steps of the systems biology approach: component identification and modeling; system perturbation and monitoring; model refining; and model testing. In addition to highthroughput monitoring and reverse-genetics tools, the volume and nature of the data now being generated necessitate the integration of computational modeling and testing. The basic premise is that the identification of a parts list, combined with the monitoring and modeling of the dynamic behavior of those parts, will offer better models of biological systems. The challenge of systems biology. The application of a systems theoretic analysis may dramatically improve our understanding of many biological processes such as morphogenesis, pathogenesis, cognition and ecology. There is no question that these all may be complex systems that preclude full understanding by the reductionistic paradigm. But this new incarnation of systems biology faces a number of challenges in the coming years. The first is technological: to move beyond a focus on transcriptional profiling and protein–protein interactions. A more complete understanding of system dynamics requires knowledge from other parts of the system, such as metabolites, as well as other forms of network

modulation, such as posttranscriptional regulation. The second, and perhaps most vexing hurdle, is data sharing. If different datasets are to be integrated in a meaningful way, it will be necessary to not only improve annotations, but also to increase standardization in lab procedures. It is axiomatic in molecular biology that minor variations in procedures, reagents or environment can have a dramatic effect on biological systems. It is thus incumbent on systems biology researchers to communicate more effectively about system parameters that may be relevant to the ultimate model. Lastly, this approach requires expertise in molecular biology, mathematics, genomics, computer science and systems theory. As it is improbable that effective progress in the field will be made by individual researchers with all of these skills, tight collaborations, and more interdisciplinary training, are essential for systems biology to be productive. Moreover, this cooperation must be greater than the sum of its PIs by not only asking molecular biological questions from a systems biology mindset, but by investigating systems biology problems with all of the skills of the individual researchers involved. The challenges are great, but so are the prospects for a new understanding of what is life. Where can I find out more? Alon, U., Surette, M.G., Barkai, N., and Leibler, S. (1999). Robustness in bacterial chemotaxis. Nature 397, 168–171. Ideker, T., Galitski T. and Hood, L. (2001). A new approach to decoding life: Systems biology. Annu. Rev. Genomics Hum. Genet. 2, 343–372. Kauffman, S.A. (1993). The Origins of Order. (Oxford University Press). Kitano, H. (2002). Systems biology: A brief overview. Science 295, 1662–1664. Ge, H., Walhout, A.J.M., and Vidal, M. (2003). Integrating ‘omic’ information: a bridge between genomics and systems biology. Trends Genet. 19, 551–560. Pennisi, E. (2003). Tracing life’s circuitry. Science 302, 1646–1649. Biology Department, Duke University, Box 91000, Durham, North Carolina 27708, USA. E-mail: [email protected]

Quick guide

Medicago truncatula Gregory D. May and Richard A. Dixon

What is M. truncatula? Commonly known as ‘barrel medic’ because of the shape of its seed pods, M. truncatula is an omni-Mediterranean species grown as an annual forage legume. It is a near relative of alfalfa — the world’s economically most important forage legume. The plant is selffertile and its genome, unlike the complex ones of other legume species, is diploid (with just eight pairs of homologous chromosomes). What attributes are unique to legume species? Legumes are unique among crop species in their ability to fix atmospheric nitrogen through symbiotic relationships with bacteria of the genus Rhizobium. This ability reduces the dependence on agricultural chemical inputs and promotes soil fertility. Like many other families of plants, but unlike the Brassicaeae that includes the model plant Arabidopsis thaliana, legumes also form symbiotic relationships with mycorrhizal fungi that assist the plant in uptake of phosphate. Legumes have evolved a complex assortment of natural products, involved in both the establishment of symbiosis and in defense. These include various flavonoids, isoflavonoids and triterpene saponins, some of which are believed to benefit human health. Why has M. truncatula been chosen as a model legume? As a bona fide forage crop, M. truncatula is an excellent subject of studies on forage quality traits such as digestibility, nutritional value, palatability and silage properties. In addition to its small genome size and simple

Magazine R181

truncatula. A series of forward and reverse genetics methodologies and populations have also been established. Extensive bioinformatics resources also exist which facilitate data comparison within M. truncatula and leverage the data available for this model species to other agronomically important legume crop species.

Figure 1. The model legume Medicago truncatula. The shape of M. truncatula seed pods (top right) lead to its common name of barrel medic.

genetics, M. truncatula harbors several attributes that make it attractive as a model species. It has a short seed-to-seed generation time and abundant seed set. Extensive collections of M. truncatula ecotypes — natural geographic variants — exist, and mutant collections are readily produced by physical (fast neutron) and transposonmediated mutagenesis. M. truncatula is also host to Sinorhizobium meliloti, a rhizobium species whose genome has been fully sequenced. What resources are available for M. truncatula research? With almost 200,000 expressed sequence tags (ESTs) and greater than 85 Mb of genome sequence currently deposited in publicly available databases, legume researchers have access to a rich data set that outlines the gene content of M. truncatula. In addition, communitystandardized M. truncatula DNA microarrays are available which facilitate near global transcript profiling. Many genes in M. truncatula have over 98% sequence identity with their orthologs in alfalfa, and so the M. truncatula microarrays can also be used for expression profiling of alfalfa. Collaborative research programs are underway that detail the transcript, protein and metabolite profiles of M.

Is there a M. truncatula genome project? After Arabidopsis and rice, M. truncatula will be the next plant to have its genome completely sequenced. An international Medicago sequencing project has begun that involves laboratories in the US, the UK and France. It is anticipated that sequencing the Medicago genome will be completed within the next three years. The genome sequence of M. truncatula will serve as a basis for structural genomics comparisons with other legume species such as alfalfa and soybean. Where can I find out more about M. truncatula? Cook, D.R. (1999). Medicago truncatula – a model in the making! Curr. Opin. Plant Biol. 2, 301–304. Young, N.D., Mudge, J., and Ellis, T.N. (2003). Legume genomes: more than peas in a pod. Curr. Opin. Plant Biol. 6, 199–204. Dixon, R.A., and Sumner, L.W. (2003). Legume natural products. Understanding and manipulating complex pathways for human and animal health. Plant Physiol. 131, 878–885. The Center for Medicago Genomics Research: www.noble.org/medicago/index.html The Consensus Legume Database. www.legumes.org Medicago genome project at the University of Oklahoma. www.genome.ou.edu/medicago.ht ml The Legume Information System. www.comparative-legumes.org Toulouse, C.N.R.S.-I.N.R.A. http://medicago.toulouse.inra.fr/Mt /EST/ The Medicago truncatula Gene Index. www.tigr.org/tdb/tgi/mtgi/ Medicago Bioinformatics at the University of California – Davis. http://medicago.plantpath.ucdavis. edu/ Plant Biology Division, The Samuel Roberts Noble Foundation, 2510 Sam Noble, Parkway, Ardmore, Oklahoma 73401, USA. E-mail: [email protected]

Correspondence

‘Spalog’ and ‘sequelog’: neutral terms for spatial and sequence similarity Alexander Varshavsky Similarities amongst sequences or three-dimensional (3-D) structures and conjectures based on similarities are a major topic of molecular biology and related fields. Therefore it is striking that there are presently no terms that denote a sequence or a 3-D structure that is similar to another sequence or 3-D structure without implying anything at all about evolutionary relatedness or biological functions. The lack of such neutral terms for denoting similarity is one reason for the widespread use of the terms ‘homolog’, ‘ortholog’ and ‘paralog’. The first term is more than a century old and the other two were proposed long before the advent of extensive sequence comparisons [1]. To state that a gene or a protein A is a homolog of B implies that A and B are related through common descent, a proposition that needs to be proven in most cases [2]. In addition, two sequences can be 37% identical, but they cannot be 37% homologous — they are either homologous or not. The frequent unsuitability of the term ‘homolog’ in the context of similarity was pointed out repeatedly [2,3], but the literature is still rife with this misuse, in part because proper neutral terms simply do not exist. The disposition can be also difficult with the terms ‘ortholog’ and ‘paralog’. Orthologs are two homologous sequences that diverged following speciation, such that the common precursor of two sequences was harboured by the last common ancestor of the two species. Paralogs, by contrast, are two homologous